Geekbench ML Score

602
TensorFlow Lite CPU Inference Score
Geekbench ML 0.5.2 for iOS AArch64

Result Information

Upload Date September 30 2022 01:31 AM
Views 1

System Information

System Information
Operating System iOS 15.7
Model iPad Air (3rd generation)
Model IDiPad11,4
Motherboard J218AP
CPU Information
Name Apple A12 Bionic
Topology 1 Processor, 6 Cores
Identifier ARM
Base Frequency 2.49 GHz
L1 Instruction Cache48.0 KB x 1
L1 Data Cache32.0 KB x 1
L2 Cache2.00 MB x 1
Memory Information
Size 2.84 GB
Inference Information
Framework TensorFlow Lite
Backend CPU

Inference Performance

Workload Accuracy Score
Image Classification (F32)
100% 481
51.6 IPS
 
Image Classification (F16)
100% 480
51.5 IPS
 
Image Classification (I8)
81% 289
31.0 IPS
 
Image Segmentation (F32)
100% 777
4.80 IPS
 
Image Segmentation (F16)
100% 768
4.75 IPS
 
Image Segmentation (I8)
89% 980
6.05 IPS
 
Pose Estimation (F32)
100% 257
1.04 IPS
 
Pose Estimation (F16)
100% 253
1.03 IPS
 
Pose Estimation (I8)
80% 275
1.11 IPS
 
Object Detection (F32)
100% 711
25.0 IPS
 
Object Detection (F16)
100% 708
24.9 IPS
 
Object Detection (I8)
52% 559
19.6 IPS
 
Face Detection (F32)
100% 897
10.3 IPS
 
Face Detection (F16)
100% 889
10.2 IPS
 
Face Detection (I8)
75% 648
7.44 IPS
 
Text Classification (F32)
100% 805
318.3 IPS
 
Text Classification (F16)
100% 802
317.4 IPS
 
Machine Translation (F32)
100% 1001
154.2 IPS
 
Machine Translation (F16)
100% 1003
154.6 IPS
 

Compare

Set Baseline

Share

Geekbench ML now available for download